Amination of Bis(trimethylsilyl)-1,2-bisketene with Secondary Amines: Formation of Aminodihydrofuranones

Article abstract of DOI:10.1021/jo000508k

The bisketene (Me₃SiC=C=O)₂ (3) reacts rapidly with 1 equiv of secondary amines to form aminodihydrofuranones 11 as the only observable products. This is in contrast to previous studies (J. Org. Chem. 1999, 64, 4690) of the reactions

Full text of DOI:10.1021/jo000508k

5676
J . Org. Chem. 2000, 65, 5676-5679
Am in a tion of Bis(tr im eth ylsilyl)-1,2-bisk eten e w ith Secon d a r y
Am in es: F or m a tion of Am in od ih yd r ofu r a n on es
Annette D. Allen, Wen-wei Huang, Patrick A. Moore, Adel Rafai Far, and Thomas T. Tidwell*
Department of Chemistry, University of Toronto, Toronto, Ontario M5S 3H6, Canada
ttidwell@chem.utoronto.ca
Received April 4, 2000
The bisketene (Me3SiCdCdO)2 (3) reacts rapidly with 1 equiv of secondary amines to form
aminodihydrofuranones 11 as the only observable products. This is in contrast to previous studies
(J . Org. Chem. 1999, 64, 4690) of the reactions of 3 with primary amines in which 3 with 1 equiv
of amine gives ketenyl amides 4, which slowly cyclize to succinimides 7. The kinetics of the reaction
of 3 with morpholine obeyed a rate law with the term [morpholine]2, consistent with rate-limiting
formation of the enol amide 14 with catalysis by a second amine molecule. The subsequent formation
of 11 is attributed to hindrance of ketonization of intermediate enol amides 14. The furanones 11
react with Me3SiOTf to form silyloxyfurans 16, and these react with diethyl diazodicarboxylate,
forming maleamide derivatives 17.
The dehydration of the monoamides of dicarboxylic
acids (amic acids 1) is a process which has been known
for over a century to give either imides or cyclic isoimides
(2, eq 1).1 These latter compounds are versatile sub-
stances that react with both nucleophiles and electro-
philes, and the isoimide of phthalic acid has been
proposed as a more transient alternative to phthalimides
in the protection of amines.1e
The reactions of ketenes with amines have been of
recent interest,2 and for 1,2-bisketene 3 (which has been
demonstrated to have a twisted conformation as shown)
with amines it was found2f that the reaction with 1 equiv
of a primary amine gave complete conversion to isolable
ketenyl amide 4, which reacted with a second equivalent
of amine to give bisamide 5 (eq 2), while reaction of 4
with alcohols gave mixed ester amides 6 (eq 3).2 Upon
standing, the ketenyl amides 4 cyclized to succinamides
7, and no evidence for isoimides analogous to 2 was
observed.2f
with amines,2e were fit by the rate law of eq 4, derived
from the mechanism of eqs 5 and 6.2e,f The formation of
the enol amide was in accord with theoretical calcula-
tions2a,b and direct observation by time-resolved IR.2d
The kinetics of the reaction of 3 with n-BuNH2 and CF3-
CH2NH2,2f as well as the reactions of PhSiMe2CHdCdO
(1) (a) Roderick, W. R.; Bhatia, P. L. J . Org. Chem. 1963, 28, 2018-
2024. (b) Boyd, G. V.; Monteil, R. L. J . Chem. Soc., Perkin Trans. 1
1978, 1338-1350. (c) Boyd, G. V.; Monteil, R. L.; Lindley, P. F.;
Mahmoud, M. M. J . Chem. Soc., Perkin Trans. 1 1978, 1351-1360.
(d) Baydar, A. E.; Boyd, G. V. J . Chem. Soc., Perkin Trans. 1 1978,
1360-1366. (e) Kukolja, S.; Lammert, S. R. J . Am. Chem. Soc. 1975,
97, 5582-5583.
However, for the reaction of 4 (R ) CF3CH2) with CF3-
CH2NH2,2f and for the reaction of t-Bu(i-Pr)CdCdO with
n-BuNH2,2e the kinetics were fit better by eq 7, derived
from the mechanism of eq 8.
(2) (a) Sung, K.; Tidwell, T. T. J . Am. Chem. Soc. 1998, 120, 3043-
3048. (b) Raspoet, G.; Nguyen, M. T.; Kelly, S.; Hegarty, A. F. J . Org.
Chem. 1998, 63, 9669-9677. (c) Andraos, J .; Kresge, A. J . J . Am. Chem.
Soc. 1992, 114, 5643-5646. (d) Wagner, B. D.; Arnold, B. R.; Brown,
G. W.; Lusztyk, J . J . Am. Chem. Soc. 1998, 120, 1827-1834. (e) Allen,
A. D.; Tidwell, T. T. J . Org. Chem. 1999, 64, 266-271. (f) Allen, A. D.;
Moore, P. A.; Missiha, A. Tidwell, T. T. J . Org. Chem. 1999, 64, 4690-
4696. (g) Huang, W.-w.; Moore, P. A.; Rafai Far, A.; Tidwell, T. T. J .
Org. Chem. 2000, 65, 3877-3879.
Cyclization has, however, been reported for reaction
of bisketene 3 with alcohols catalyzed by chelated Cu(II)
10.1021/jo000508k CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/15/2000

Bis(trimethylsilyl)-1,2-bisketene-2° Amine Amination
J . Org. Chem., Vol. 65, No. 18, 2000 5677
catalysts, giving efficient conversion to 5-alkoxydihydro-
furanones 8 (eq 9).3a,b Some precedent for cyclization in
a bisketene reaction with an amine was obtained from
the reaction of ketene 9 with pyridine (eq 10).3c The UV
and IR bands due to 9 disappeared, and a new UV
absorption at 590 nm appeared which was ascribed to
the ylide 9a ,3c but no IR absorption assigned to 9a was
observed, so this assignment is not secure. We now report
the unexpected finding that reaction of 3 with secondary
amines proceeds with efficient cyclization to isoimides.
F igu r e 1. Rate constants for reaction of 3 versus [morpholine]
and fit by eq 7.
Resu lts a n d Discu ssion
Reaction of 3 with 1 equiv of secondary amines 10a -d
at room temperature for 30 min gave the 5-amino-3,4-
bis(trimethylsilyl)dihydrofuran-2-ones 11 in 97-99% yields
(eq 11). These products were clearly identified from their
studied, and these rate laws could be derived from
reasonable mechanisms (eqs 5, 8) that are consistent with
theoretical studies.2a,b
For the reaction of 3 with morpholine, the most
satisfactory fit (Figure 1) of the data was found with the
rate law of eq 7, with values of kc ) (3.21 ( 0.23) × 104
and kd ) (3.53 ( 0.44) × 102 M-1 s-1. An almost
superimposable fit (Figure 2) was obtained with the rate
law of eq 14, which is derived for the mechanism of eq 8
with the additional step of eq 15 with values ke ) (2.35
( 0.40) × 104, kf ) (5.50 ( 2.53), and kg ) (2.22 ( 0.63)
× 102 M-1 s-1. The rate law of eq 7 is deemed more
satisfactory because of its greater simplicity and smaller
standard errors, but the latter cannot be completely ruled
out. The full rate data are given in the Supporting
Information.
spectral properties, but unlike the alkoxyfuranones 83a,b
they were rather sensitive and decomposed upon stand-
ing. Reaction of 3 with 1 equiv of i-Pr2NH (10e) was much
slower than for the other amines, and the product 11e
was quite unstable and could only be observed in solution.
The reaction of 3 with 2 equiv of Et2NH led to the
partially desilylated succinamide 12 (eq 12). Such R-si-
lylated carbonyl compounds are sensitive to desilylation.
This evidently arose from attack of the amine on the
carbonyl of the intermediate lactone 11, as the reaction
of the furanone 11a with benzylamine led to ring opening
with formation of a partially desilyated succinamide,
characterized as 13 after desilylation (eq 13).
Kinetic studies of the reaction of 3 with morpholine in
CH3CN at 25 °C were carried out and showed a higher
than first-order dependence of the rate constants on
[morpholine]. Similar behavior was observed previously
in the reactions of primary amines with monoketenes2e
and with 3,2f and a number of different rate laws were
tested to ascertain the best fit of the data.2e,f As noted
above, the particular rate law followed, such as those
shown in eqs 4 and 7, depended on the particular ketene
Because of the change in rate law of 3 for morpholine
compared to that of primary amines,2f the ratio kn-BuNH
/
2
kmorpholine for kobs is 1/50 at [amine] ) 4 × 10-3 M whereas
for PhMe2SiCHdCdO which reacts with both amines
with the rate law of eq 4 the ratio kn-BuNH /kmorpholine for
2
(3) (a) Dejmek, M.; M.; Selke, R. Angew. Chem., Int. Ed. 1998, 37,
1540-1542. (b) Dejmek, M. M.; Selke, R. Synlett 2000, 13-21. (c) Oishi,
S.; Ozaki, J . Chem. Lett. 1998, 1071.
kobs is 7.4 at [amine] ) 4.0 × 10-3 M.2e The explanation
for the absence of observable ketenyl amides, and for the

5678 J . Org. Chem., Vol. 65, No. 18, 2000
Allen et al.
Sch em e 1
F igu r e 2. Rate constants for reaction of 3 versus [morpholine]
and fit by eq 14.
would be enhanced in a ketenyl amide formed from a
secondary amine. Another alternative involving direct
cyclization of 3 to 11 upon amination involving two amine
molecules would appear to be disfavored by more steri-
cally demanding secondary amines.
occurrence of cyclization to isoimides (eq 11), may be
found from the mechanistic details of the amination of
ketenes and bisketenes. The kinetic data lead to the
preferred mechanism of eq 8, in which a second molecule
of amine is involved in the rate-limiting step. For the
reaction of 3 with morpholine, formation of the amine-
complexed enol amide 14 is thus indicated to be rate
limiting, and this is then converted to the isoimide 11
(eq 16). Thus, the mechanism of eq 8 involves formation
The recent application4 of 2-silyloxyfurans in vinyl-
ogous aldol and Mannich-type reactions suggested that
the furanones 11 could be converted to analogous deriva-
tives. The conversion of R-silylated carbonyl compounds
to the corresponding silyl enol ethers may be effected by
the addition of a catalytic amount of Me3SiOTf,5 and this
procedure gave rapid and quantitative conversion of
11a ,b to the furans 16a ,b, as monitored by 1H NMR
(Scheme 1). These products were too sensitive for isola-
tion but upon reaction with diethyl diazodicarboxylate
(DEAD) gave conversion to the ring-opened products
17a ,b. The one-pot conversions of 11c,d to 17c,d were
also carried out. The structure of product 17a was
confirmed by an X-ray determination. The reactions may
be envisaged as involving electrophilic attack at the Me3-
SiO-substituted position of the furan followed by ring
opening (Scheme 1). Initial formation of a [4 + 2]
cycloadduct of 16 and DEAD followed by rearrangement
was not observed but cannot be excluded.
In summary, the reactions of bisketene 3 with 1 equiv
of secondary amines form isoimides 11 (eq 11), in contrast
to the reaction of primary amines, which leads to ketenyl
amides 4 (eq 2).2f A mechanism (eqs 8, 16) consistent with
the theoretical,2a,b product, and kinetic studies is pre-
sented. The conversion of the isoimides 11 to furans 16
and their reaction with diethyl azodicarboxylate is dem-
onstrated.
of an initial hydrogen-bonded complex which reacts with
a second amine, forming 14 in the rate-limiting step, and
this forms 11 (eq 16). Formation of a ketenyl amide would
also proceed through the formation of enol amide 14, but
this would require catalysis by another amine molecule
to undergo ketonization to the amide. The involvement
of enol amides in ketene amination is supported by both
theoretical studies2a,b and direct observation,2d and fur-
ther conversion to the amide is amine catalyzed.2d In the
reaction of 3 with a secondary amine, this would involve
the conversion of complexed enol amide 14 to transition
structure 15, which would involve significant steric
crowding (eq 16). Evidently cyclization of enol amide 14
to isoimide 11 becomes competitive when the ketoniza-
tion is slow. An alternative pathway for the formation of
11 would be rapid conversion of an unobserved ketenyl
amide analogous to 4 from the secondary amine, but we
see no apparent reason that cyclization to the isoimide
Exp er im en ta l Section
Dichloromethane was dried over 4 Å molecular sieves 24 h
prior to use. CDCl3 was dried over anhydrous potassium
(4) (a) Rassu, G.; Zanardi, F.; Battaistini, L.; Casiraghi, G. Synlett
1999, 1333-1350. (b) Martin, S. F.; Barr, K. J .; Smith, D. W.; Bur, S.
K. J . Am. Chem. Soc. 1999, 121, 6990-6997. (c) Keay, B. A. Chem.
Soc. Rev. 1999, 28, 209-215.
(5) (a) Emde, H.; Go¨tz, A.; Hofmann, K.; Simchen, G. Liebigs Ann.
1981, 1643-1657. (b) Yamamoto, Y.; Ohdoi, K.; Nakatani, M.; Akiba,
K.-y. Chem. Lett. 1984, 1967-1968.

Bis(trimethylsilyl)-1,2-bisketene-2° Amine Amination
J . Org. Chem., Vol. 65, No. 18, 2000 5679
carbonate overnight. Trimethylsilyl triflate was freshly dis-
tilled from calcium hydride. Morpholine, piperidine, and
diisopropylamine were distilled from calcium hydride. Bis-
ketene 3 was made by injection of a solution of 3,4-bis-
(trimethylsilyl)cyclobutenedione in pentane (approximately
300 µL per 200 mg) through a gas chromatograph with an
injector temperature of 200 °C and an OV-17 column heated
at 120 °C. All other reagents were used directly from com-
mercially available sources. All the reactions were performed
under a positive pressure of argon.
mmol of a secondary amine. After stirring for 30 min, 10 µL
(0.05 mmol) of trimethylsilyl triflate was added, giving a purple
solution. The reactions with BnNHCH3 and morpholine were
checked by 1H NMR and were complete after 15 min. The
furans thus prepared were then used directly in the next step.
16a : 1H NMR δ 0.18 (s, 9), 0.32 (s, 9), 2.60 (s, 3), 4.04 (s, 2),
4.87 (s, 1), 7.32 (m, 5); 13C NMR δ -0.6, -0.2, 41.8, 60.7, 86.2,
107.9, 127.0, 129.2, 128.2, 128.4, 128.5, 128.8, 138.5, 152.2,
153.7.
16b: 1H NMR δ 0.17 (s, 9), 0.28 (s, 9), 2.94 (t, J ) 4.7 Hz,
4), 3.73 (t, J ) 4.6 Hz, 4), 4.84 (s, 1).
Rea ction of Bisk eten e 3 w ith Secon d a r y Am in es. To a
solution of bisketene (100 mg, 0.44 mmol) in 2 mL of CH2Cl2
or CDCl3 at 25 °C was added 0.44 mmol of the amine. After
30 min, the solvent was removed in vacuo to give the product
as an oil. The products 11a -d were obtained in 97-99% crude
yields but are rather sensitive, and except for 11c they were
not purified and were characterized as described. The product
11e formed slowly and was quite unstable, so the same
reaction as above was performed in CDCl3 for 6 h and the
product was not isolated but was analyzed by NMR directly.
11a : 1H NMR (CDCl3) δ 0.16 (s, 9), 0.164 (s, 9), 2.67 (s, 3),
3.04 (s, 1), 4.06 (d, 1, J ) 14.3 Hz), 4.27 (d, 1, J ) 14.3 Hz),
7.24-7.36 (m, 5); 13C NMR (CDCl3) δ -2.6, 0.5, 39.1, 44.9, 57.6,
91.3, 127.4, 128.4, 128.4, 137.2, 160.4, 178.6; IR (neat) 1762,
1618 cm-1; EIMS m/z 347 (M+, 70), 256 (100), 73 (89); HRMS
m/z calcd for C18H29NO2Si2 347.1737, found 347.1751.
Rea ction of F u r a n s w ith DEAD. To the furan solutions
prepared from 100 mg of bisketene 3 was added dropwise 69
µL (0.44 mmol) of DEAD. The resulting light brown solutions
were stirred for 15 min, 2 mL of methanol was added, and the
mixtures were stirred for 30 min. The solvents were then
removed in vacuo, and the resulting products were purified
by flash chromatography on silica gel, using the indicated
eluent and giving the yields shown. In the 1H NMR spectra
the vinylic proton could not be detected.
17a : eluent EtOAc (75%); mp 106-108 °C; 1H NMR δ 0.23
(s, 9), 1.27 (m, 3), 1.31 (t, J ) 7.1 Hz, 3), 2.72 (s, 3), 4.23 (m,
6), 7.01 (broad s, 1), 7.27 (m, 5); 13C NMR δ -1.9, 14.1, 14.3,
34.6, 49.7, 53.9, 62.3, 64.0, 127.2, 127.4, 128.4, 128.5, 137.0,
153.0, 155.4, 157.6, 164.2, 171.5; EIMS m/z 450 (MH+, <1),
91 (100), 73 (29); HRMS m/z calcd for C21H32N3O6Si (MH+)
450.2060, found 450.2067; IR (CDCl3) 1748 (broad but strong)
11b: 1H NMR (CDCl3) δ 0.15 (s, 9), 0.17 (s, 9), 2.98 (m, 5),
3.62 (m, 4). 13C NMR (CDCl3) δ -2.6, 0.0, 44.7, 50.6, 66.5, 99.7,
159.8, 178.6; EIMS m/z 313 (M+, 61), 73 (100); HRMS m/z calcd
for C14H27NO3Si2 313.1530, found 313.1529; IR (CDCl3) 1748,
1616 cm-1
.
1
17b: eluent 1/4 hexanes/EtOAc (70%); H NMR (CDCl3) δ
0.23 (s, 9), 1.28 (m, 3), 1.32 (t, J ) 7.1 Hz, 3), 3.64 (broad m,
8), 4.20 (q, J ) 7.1 Hz, 2), 4.30 (q, J ) 7.1 Hz, 2), 7.05 (broad
s, 1); 13C NMR (CDCl3) δ -1.9, 14.1, 14.4, 41.4, 46.4, 62.5, 64.1,
66.3, 66.5, 128.4, 153.0, 155.4, 157.8, 164.1, 170.3; EIMS m/z
415 (M+, <1%), 240 (100), 73 (44); HRMS m/z calcd for
1638 (broad and strong) cm-1
.
11c: Rea ction of Bisk eten e 3 w ith (R)-(+)-N,r-Dim -
eth ylben zyla m in e (10c). To bisketene 3 (0.133 g, 0.59 mmol)
in 10 mL of CH2Cl2 stirred at -78 °C was added amine 10c
(0.080 g, 0.58 mmol) in 2 mL of CH2Cl2. The solution was
stirred for 1 h at -78 °C, and then the solvent was evaporated,
C
17H30N3O7Si (MH+) 416.1853, found 416.1837; IR (CDCl3)
1752 (broad but strong), 1616 cm-1
.
1
1
giving crude 11c as an oil (99%) which by H NMR consisted
17c: eluent 1/1 hexanes/EtOAc (51%); H NMR (CDCl3) δ
0.24 (s, 9), 1.29 (m, 3), 1.31 (t, J ) 7.1 Hz, 3), 1.51 (d, J ) 7.0
Hz, 3), 2.47 (s, 3), 4.19 (m, 3), 4.29 (q, J ) 7.1 Hz, 2), 7.00
(broad s, 1), 7.25 (m, 5); 13C NMR (CDCl3) δ -1.8, 14.1, 14.2,
29.5, 49.7, 62.4, 63.9, 122.8, 127.1, 127.7, 128.2, 150.4, 153.1,
155.5, 164.1, 171.2; EIMS m/z 464 (MH+, 57), 134 (100), 73
(30); HRMS m/z calcd for C22H34N3O6Si (MH+) 464.2217, found
of two diastereomers in a 3/1 ratio. The major isomer was
obtained by recrystallation of the crude product from ether at
-20 °C as a white crystalline solid: 1H NMR (CDCl3) δ 0.12
(s, 9), 0.14 (s, 9), 1.36 (d, 3, J ) 6.8 Hz, CH3), 2.42 (s, 3), 2.98
(s, 1), 4.22 (q, 1, J ) 6.8 Hz), 7.20-7.34 (m, 5); 13C NMR
(CDCl3) δ -1.9, 0.4, 20.4, 38.3, 44.7, 60.3, 98.1, 127.2, 128.4,
142.8, 159.8, 179.2 (one phenyl carbon not seen); IR (CDCl3)
1746, 1637, 1607 cm-1; EIMS m/z 361 (M+, 14), 73 (100);
HRMS m/z calcd for C19H31NO2Si2 361.1893, found 361.1876.
11d : 1H NMR (CDCl3) δ 0.14 (s, 9), 0.15 (s, 9), 1.55 (m, 6),
2.93 (m, 5); 13C NMR (CDCl3) δ -2.6, -0.1, 23.5, 25.6, 44.6,
51.5, 97.2, 161.3, 179.1; EIMS m/z 311 (M+, 100), 73 (83);
HRMS m/z calcd for C15H29NO2Si2 311.1737, found 311.1730;
464.2207; IR (CDCl3) 1744 (broad but strong), 1616 cm-1
.
1
17d : eluent 1/1 hexanes/EtOAc (51%); H NMR (CDCl3) δ
0.22 (s, 9), 1.27 (m, 3), 1.32 (t, J ) 7.2 Hz, 3), 1.59 (broad m,
6), 3.23 (broad t, J ) 4.8 Hz, 4), 4.19 (q, J ) 7.2 Hz, 2), 4.30
(q, J ) 7.2 Hz, 2), 6.96 (broad s, 1); 13C NMR (CDCl3) δ -1.8,
14.1, 14.3, 24.5, 25.2, 25.6, 41.8. 46.9, 62.4, 64.0, 127.6, 153.0,
155.4, 164.3, 169.9, 179.3; EIMS m/z 414 (MH+, 37), 238 (100),
73 (15); HRMS m/z calcd for C18H32N3O6Si (MH+) 414.2060,
IR (CDCl3) 1750, 1637 cm-1
.
11e: 1H NMR (CDCl3) δ 0.18 (s, 9), 0.20 (s, 9), 1.10 [m made
of 1.10 (d, J ) 7.3 Hz, 6) and 1.11 (d, J ) 6.6 Hz, 6)], 2.98 (s,
1), 3.34 (septet, J ) 6.6 Hz, 2); 13C NMR (CDCl3) δ -2.4, -0.2,
21.4, 21.5, 33.2, 51.8, 84.9, 173.0 (one carbon missing); IR
found 414.2044; IR (CDCl3) 1748 (broad but strong), 1616 cm-1
.
Kin etic P r oced u r e. Rate constants were obtained by
injecting aliquots of freshly prepared solutions of 3 (0.229 or
0.257 M) in CH3CN into 2 mL solutions of morpholine [(0.293
to 5.85) × 10-3 M] in CH3CN in a UV cell equipped with a
magnetic stirrer and observing the decrease in the ketene
absorption at 385 nm using a Perkin-Elmer Lambda 12
instrument. Initial concentrations of 3 ranged from 6.41 × 10-4
to 0.286 × 10-4 M. The program SigmaPlot was used to fit
the kinetics, using a statistical error weighting program. A
weighting of 1/kobs gave the best fit of the data. A variety of
other rate laws were tested, as we have done previously.2e
(CDCl3) 1782, 1634 cm-1
.
Rea ction of 11a w ith Ben zyla m in e. To a solution of 11a
(151 mg, 0.43 mmol) in 2 mL of CH2Cl2 was added 48 µL (47
mg, 0.44 mmol) of benzylamine. The mixture was stirred at
room temperature for 1 h. The solvents were removed in vacuo,
the residue was taken up in 2 mL of methanol, and 200 µL of
TFA were added. After 30 min, the volatiles were removed in
vacuo, and the residue was purified by chromatography (2/8
methanol/CH2Cl2), to give 13 (103 mg, 0.33 mmol, 76%): 1H
NMR 2.60 (m, 2), 2.73 (broad t, J ) 6.2 Hz, 2), 2.89 (s, 51% of
3), 2.91 (s, 49% of 3), 4.41 (m, 2), 4.53 (d, J ) 6.2 Hz, 2), 6.73
(broad s, 42% of 1), 6.81 (broad s, 58% of 1), 7.24 (m, 10); 13C
NMR δ 28.7, 29.1, 31.4, 34.0, 34.7, 43.5, 51.0, 53.2, 126.4, 127.2,
127.4, 127.60, 127.61, 127.64, 127.8, 128.5, 128.6, 128.9, 137.0,
138.4, 172.6 (broad); EIMS m/z 310 (M+, 21), 91 (100); HRMS
m/z calcd for C19H22N2O2 310.1681, found 310.1670; IR (CDCl3)
Ack n ow led gm en t. Financial support by the Natu-
ral Sciences and Engineering Research Council of
Canada is gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Kinetic data, 1H
NMR spectra, and an X-ray crystallographic file on 17a in CIF
format. This material is available free of charge via the
Internet at http://pubs.acs.org.
1631 cm-1
.
P r ep a r a tion of F u r a n s. To a solution of bisketene 3 (100
mg, 0.44 mmol) in 2 mL of CH2Cl2 or CDCl3 was added 0.44
J O000508K